DK2981711T3

DK2981711T3 - PROCEDURE FOR EFFICIENCY MONITORING OF A WINDOW ENERGY PARK

Info

Publication number: DK2981711T3
Application number: DK14709552.5T
Authority: DK
Inventors: Niko Mittelmeier
Original assignee: Senvion Gmbh
Priority date: 2013-04-03
Filing date: 2014-03-07
Publication date: 2018-10-22
Also published as: EP2981711B1; EP2981711A1; ES2687867T3; WO2014161626A1; DE102013205838A1

Description

The invention relates to a method for efficiency monitoring of a wind farm.

For an operator of a wind farm it is very important to know whether the efficiency of the wind farm changes in the course of the operation of the wind farm in order to be able to perform interventions if necessary, which ensure the highest possible energy yield of the wind farm. Given the fact that a wind farm can be subject to many boundary conditions or parameters which affect the power yield or energy yield, an efficiency monitoring is accordingly complex. DE 10 2011 081 241 A1 discloses a method for determining an energy yield loss of a first and sole wind turbine of a wind farm with a plurality of wind turbines.

The object of the present invention is to specify a method for efficiency monitoring of a wind farm, which enables the efficiency of the wind farm to be reliably determined and compared so that it can be monitored.

This object is achieved by a method for efficiency monitoring of a wind farm with the following process steps: • measuring at least one wind speed and at least one wind direction during a specifiable time period, • measuring an output power of the wind farm in the specifiable time period, • associating the output power with the measured wind speed and the measured wind direction, • storing the output power as an output power value in a first power matrix, in which matrix elements are associated with various parameters comprising at least wind speeds and wind directions, as long as the output power represents a representative value for the wind farm, • providing a first counting matrix having matrix elements which are associated with the various parameters comprising at least wind speeds and wind directions, • adding a counting value to the matrix element of the first counting matrix that corresponds to the measured wind speed and the measured wind direction, • wherein the preceding process steps are carried out for a plurality of specifiable time periods in a reference time period, • wherein the following process steps are carried out in a later monitoring time period for a plurality of specifiable time periods: o measuring at least the wind speed and the wind direction during the specifiable time period, o measuring an output power of the wind farm in the specifiable time period, o providing a second counting matrix having matrix elements that are associated with the various parameters comprising at least wind speeds and wind directions, o adding a counting value to the matrix element of the second counting matrix that corresponds to the measured wind speed and the measured wind direction, • wherein, in order to monitor the efficiency, a comparison is performed between the output power values of the first power matrix weighted with the second counting matrix, and the sum of the output power values measured during the monitoring time period.

In the context of the invention a weighting of the output power values of the first output power matrix or another output power matrix means that the parameter values which are stored in a first or second counting matrix and which comprise the wind direction and wind speed, but can also include other parameters, are taken into account.

In this case, for example, the output power values of the first output power matrix, provided the output power values are, for example, averaged output powers, are multiplied by the associated value of the second counting matrix. This then results in an output power which would have been achieved in the monitoring time period if the wind farm had had the same quality of the wind farm as was the case during the reference time period.

In the output power matrix, matrix elements can be present which each represent an averaged output power. The total output power can also be stored, or else a percentage value of an output power or output power value in terms of the overall output power of the wind farm. A weighting of the power values by means of the matrix elements provided in the counting matrix, in particular the second counting matrix, is performed in different ways depending on which matrix elements are stored in the output power matrix. If an average value for an actual power is stored, the corresponding matrix element of the output power matrix is multiplied with the associated matrix element of the counting matrix.

If a total of the output powers is stored in the matrix elements of the output power matrix, the sum is first divided by the number of values stored in a first counting matrix in the appropriate matrix elements, in order then to be multiplied by the associated matrix element in the second counting matrix. If a percentage value relative to the overall output power of the wind farm is stored in the output power matrix, the same operations as described above are carried out, except that the percentage output power is also multiplied by the overall output power of the wind farm.

In the context of the present invention a wind farm comprises, in particular, a plurality of wind turbines. A measuring mast is preferably provided, which is erected in the wind farm and positioned remotely from the wind turbines. By means of the measuring mast a wind speed and a wind direction can be reliably measured. Alternatively, the wind measurement can also be performed using, in particular calibrated, nacelle anemometers .

The method is preferably applied to the entire wind farm but can also be applied only to parts of a wind farm, of course. This is particularly advantageous in the case of a wind farm with different types of wind turbine, for example, where each type is intended to be evaluated separately.

The measured wind speed is preferably corrected for the current air density by means of temperature and air pressure measurement. Methods of this kind are known in the prior art. The air humidity can also be advantageously taken into account to increase the accuracy in a wind speed correction.

The specifiable time period in the context of the invention can preferably be in the range from 2 minutes to 20 minutes, in particular with a value of 10 minutes. Other time periods can also be provided. In the specified time period, the speed and the wind direction are measured, together with an output power of the wind farm. The power of the wind farm can be, for example, the sum of the measured output powers of each wind turbine that is in operation, or the power output by the wind farm into a network. A counting matrix is preferably a parameter distribution matrix, in particular a wind distribution matrix.

The process steps are preferably carried out for a representative plurality of specifiable time periods in the reference time period and the monitoring time period. A representative plurality of time periods means, in particular, that the totality of the time periods is representative of the reference time period or monitoring time period. In particular, this means that the distribution of the parameter or parameters, for example the wind direction and/or wind strength, in the plurality of time periods taken as the basis should be representative of the entire reference or monitoring time period. For example, an annual wind distribution can be reproduced with the desired accuracy. If 100% accuracy is required, the measurement is required to be carried out continuously.

Preferably, during the monitoring time period a second output power matrix is formed by the output powers measured in the monitoring time period being associated with the wind speeds and wind directions measured in the respective time period and the output powers being stored in the second output power matrix as output power values, wherein different parameters, comprising at least wind speeds and wind directions, are associated to the output power values in the second output power matrix.

Preferably, the method is executed and/or the corresponding measurements recorded only when all wind turbines of the wind farm are in operation. Alternatively, care must be taken to ensure that the configuration during the reference time period also corresponds to the configuration during the monitoring time period. When assigning the output power to the measured wind speed and the measured wind direction, the measured output power at the measured wind speed and at the measured wind direction is associated with these boundary parameters accordingly. This output power value is then subsequently stored in a first output power matrix. The output power matrix then has matrix elements that are associated with the parameters for wind speed and wind direction. Thus, in this case the matrix can be two tuples or a matrix with two dimensions, one dimension of which is the wind direction and the other dimension the wind speed. For example, wind speeds from one meter per second up to 25 metres per second can be provided, in increments of one metre per second, and thus 26 values for the wind speed can be provided in the wind speed dimension or the wind speed tuple of the output power matrix. Accordingly, the wind direction can also be divided into sections, for example, into five-degree classifications or in 10-degree or 15-degree classifications or other classifications which make sense for the respective wind farm. For example, with ten-degree classifications, 36 different wind directions or wind direction values are therefore possible. In the above examples, for example, this results in a matrix with 26 times 36 values in which the corresponding output power values are stored.

The output power values are preferably stored in the reference time period only if they are representative of the wind farm. This also means, in particular, a system availability within the usual parameters. If, for example due to an extremely rare weather event, a large number of wind turbines are iced up and are not delivering their usual output power, such a state is preferably not stored in the output power matrix of the reference time period or the monitoring time period. A first counting matrix is also provided, which also has the appropriate number of dimensions in relation to the various parameters such as wind speed and wind direction. In the above example, 26 times 35 matrix elements are also obtained, which can be described in the counting matrix. In the case of an appropriately measured wind speed and an appropriately measured wind direction a counting value, such as a one, will therefore be added to the matrix element which corresponds to these measured parameters (wind speed and wind direction). Therefore, if a wind with, for example, a wind speed of 5 meters per second from a wind direction of 200° was measured for the first time, the matrix element associated with this data receives the value one. If this happens for a third time, the associated matrix element receives the value three. These steps are executed for a plurality of the specifiable time periods within a reference time period, which means that firstly, the matrix elements of the output power matrix are completely filled with output power values and the first counting matrix is filled in the same way. The first counting matrix can then be considered a kind of wind distribution matrix over the reference time period.

Preferably, the output power matrix and the counting matrix comprise the same parameters or an identical parameter tuple. It is also possible, however, to capture additional parameters such as the wind gradient in one of the actual matrices, for example the counting matrix. In the weighting of the output power matrix with the counting matrix all the wind gradient classes will then be grouped together. This has the advantage that in the event that a link is later found between wind gradient and output power, this can be assessed retrospectively.

The output power values that are stored in the first output power matrix are preferably corresponding average values of the output power values to be stored in the respective matrix elements. For example, if output powers for a specific wind direction and specific wind speed are recorded five hundred times during the reference time period, the output power values that are stored in the corresponding matrix element of the first output power matrix are the sum of the five hundred measured output power values divided by five hundred.

It is also possible, however, to store the sum of the measured output power values as the output power value, and later to perform an appropriate normalization of the output power values using the corresponding counting values of the first counting matrix. In a further advantageous embodiment, the output power values are stored as discrete values. This advantageously enables a subsequent statistical evaluation.

The same process steps are then preferably executed in a subsequent monitoring time period, wherein a second output power matrix and a second counting matrix are formed.

For efficiency monitoring a comparison is preferably made of the output power values of the first power matrix weighted with the second counting matrix, with the second output power matrix. To this end a comparison of the energy yields is performed, for example by multiplying the first output power matrix with the second counting matrix and multiplying the second output power matrix with the second counting matrix. This is used to check the amount of overall output power or total energy yield the wind farm would have had in the monitoring time period if an output power matrix in accordance with the reference time period had prevailed. This value is accordingly compared with the actual overall output power in the monitoring time period.

The advantage of compiling the second output power matrix is that, in the event of substantial deviations in the output power values of the matrix elements of the output power matrices to be associated between the reference time period and monitoring time period, the possibility exists for the power matrices to be compared element by element in order to determine the causes of the deviations. If the deviations only occur at certain wind speeds or wind directions, this can provide important information for error diagnosis. If, on the other hand, the deviations occur uniformly over the entire time period, the cause could lie in power-independent components, such as a transformer or relay stations, or even in the measuring arrangement.

The inventive idea in this case is to measure, for example, an annual yield of a wind farm in such a way that by means of an output power matrix and a counting matrix, or wind distribution matrix, the annual yield in subsequent years is determined by means of the first output power matrix and a current wind distribution matrix, in other words a second counting matrix. This is then compared with the current energy yield or the current total power. To enable this, the matrix elements of the different matrices should preferably have an identical association for the parameters of wind speed and wind direction.

The larger the ranges which are represented in the dimensions of the matrices, that is, the greater the wind sector that is used for a measurement and the greater the wind speed variation that is added to a measurement, the fewer matrix elements occur and the better are the statistics for the respective values. At the same time, however, the accuracy of the output power measurement becomes worse, since the output power values for different wind directions and wind speeds are inherently subject to large variation. For this reason, for each wind farm the size of the respective parameter sectors of the matrices is preferably adjusted, or the number of matrix elements is preferably adjusted.

Preferably, the average values of the measured wind speed and the measured wind direction are stored.

An average value of the output power values in the respective matrix elements of the first output power matrix and the second output power matrix is preferably stored. In order to monitor the efficiency, only those output power values are taken into account which have at least n counting values in the respective counting matrix, wherein n is a natural number which is larger than 1 and is specifiable. Preferably, n is greater than or equal to five, in particular n is preferably greater than or equal to 10, more preferably n is greater than or equal to 15.

This can improve the statistics considerably and increase the reliability of the method.

In the event that the output power matrices already contain averaged output powers as output power values, in the comparison a sum is formed of the products of the output power values of the first output power matrix multiplied by the respective counting value of the second counting matrix and a sum of the products of output power values of the second output power matrix multiplied by the respective counting value of the second counting matrix. Both of these sums then represent energy yields and/or output power sums and can be directly compared with each other. A virtual "reference energy yield", which would have been produced if the wind farm had been operated in the monitoring time period with output power characteristics from the reference time period, is therefore compared with the (real) energy yield of the monitoring period which is measured as being representative.

In the event that the output power matrices contain the sums of the measured output powers as output power values, in the comparison a sum of the products of the output power values of the first output power matrix multiplied by the quotient of the respective counting value of the second counting matrix divided by the respective counting value of the first counting matrix is compared with the sum of the output power values of the second output power matrix. In this case, the output power matrix is in fact a yield matrix, since the stored output powers were actually present during the specifiable time period.

Preferably, in the comparison a sum is formed of the output power values of the first output power matrix multiplied by the respective counting value of the second counting matrix, and a sum of the output power values of the second output power matrix multiplied by the respective counting value of the second counting matrix.

Preferably, the sum of the output power values measured during the monitoring time period results from forming a sum of the output power values of the second power matrix multiplied by the corresponding counting value of the second counting matrix .

Preferably, an efficiency change is only found when a tolerance threshold is exceeded. The tolerance threshold in this case is used to absorb or allow for an uncertainty factor or statistical fluctuations.

Preferably, a wind gradient, a wind shear, a wind turbulence, an air density, an air humidity and/or an ambient temperature are measured as additional parameters. Preferably, these additional parameters or at least one other parameter are taken into account in the various matrices, i.e. the first output power matrix, the second output power matrix, the first counting matrix and the second counting matrix, as an additional dimension or an additional tuple. The more additional parameters are used, the more accurate is the efficiency monitoring, wherein it should be taken into account, however, that the values in the counting matrix become smaller the more additional parameters are taken into account. This makes the statistics worse. For this reason, an optimization is performed here in order not to take into account too many additional parameters. Which additional parameters are to be taken into account depends largely on the location of the wind farm and the external boundary conditions. In a location where very high air density fluctuations are prevalent, it will be sensible to take the air density into account. In a location which is characterized by a very large time-varying wind gradient (i.e. horizontal or vertical wind shear) , it will be sensible to take this into account.

Preferably, a voltage level in the electricity network and/or a reactive power feeding is taken into account as a further parameter. This is of interest in particular in the case where a wind farm is connected, for example, to a network in which there is a high probability that the network operator requires a corresponding amount of reactive power to be fed in to support the network. In this case, an increased reactive power in-feed affects the measured output powers, which it would be sensible to take into account. It also makes sense to provide an additional dimension in the existing matrices or another tuple, for these other parameters. Accordingly, it is preferred if the first output power matrix, the second output power matrix, the first counting matrix and the second counting matrix provide an additional dimension for an additional parameter.

The reference time period and the monitoring time period is preferably a complete meteorological year. In this case, the beginning and the end of the year can be shifted accordingly. This means, for example, a complete meteorological year can start on 01.03. and end on 28.02. or begin on 01.01. and end on 31.12. The time periods can also be integer multiples of a meteorological year; for example, 2 or 3 years.

Preferably, the measured values of the wind speeds and wind directions are sub-divided into specifiable ranges. The values of the additional parameters are preferably sub-divided into specifiable ranges.

Preferably, it is possible to omit certain parameters in the method for monitoring the efficiency of the wind farm. For example, it may be useful not to take into account certain wind direction sectors, since in corresponding wind farms a very large turbulence is to be expected from certain wind directions or wind direction sectors. Thus, for example, a corresponding output power matrix, namely a first output power matrix, which can be denoted by PRij, can have corresponding matrix elements with the indices ij, where i can be counted from 0 to 25 and j, for example, from 105 to 285°, wherein i is intended to represent the classification for the wind speed and j the classification for the wind sectors. In one example in which the wind sectors each cover 30°, 6 sectors would therefore be provided for j=105° to 285°. The wind speed is then also sub-divided into corresponding sectors, for example from 0 to 1 m/s, 1 to 2 m/s, .... 24 to 25 m/s. Reference output powers or the overall output power of the reference time period can thus be calculated as follows:

In this equation RPMT is the potential total yield of the wind farm in the monitoring time period. Nij are the respective contents of the second counting matrix.

Accordingly, an actual total yield of the wind farm for a monitoring time period can be calculated from the second output power matrix according to the following formula:

where PGij are the various output power values of the second output power matrix and N±j the counting values of the second counting matrix. A comparison of CPMT and RPMT shows how efficient the wind farm is in the monitoring time period. If CPMT should be significantly less than RPMT, appropriate measures would need to be implemented, such as the cleaning of the rotor blades or removal of vegetation in the vicinity of the wind farm, for example, trees that might have grown too tall. A yield ratio PR can be formed as follows:

With the method for monitoring the efficiency of the wind farm economic considerations could even be taken into account, such as the review of whether the wind farm is providing the operator with the output power promised by the wind farm manufacturer. A contractual penalty could also be charged, which applies in the event that the efficiency of the wind farm significantly reduces over time. An appropriate compensation payment could be implemented using the following formula :

wherein CP is the compensation payment, EPC the energy production of a monitoring time period in kWh, PR is calculated according to the above formula 3, UC is a specifiable uncertainty of the model in per cent and FI are the costs per kWh. The factor 1.28 is in principle a freely selectable factor for the uncertainties of the method, which is captured by the standard deviation.

The factor 1.28 is also known as the so-called "coverage factor", by which the confidence level is increased.

The following applies: 1.0 corresponds to 68% confidence level 2.58 corresponds to 99% confidence level 3.0 corresponds to 99.7% confidence level

Further features of the invention will become apparent from the description of embodiments in accordance with the invention in conjunction with the claims and the accompanying drawings. Embodiments according to the invention can fulfil individual characteristics or a combination of a plurality of features .

The invention is described below without limiting the general idea of the invention on the basis of exemplary embodiments and with reference to the drawings, wherein in respect of all details according to the invention which are not explained in detail in the text explicit reference is made to the drawings. Shown are :

Fig. 1 a schematic sequence of part of the method according to the invention,

Fig. 2 a schematic processing sequence of part of the method according to the invention,

Fig. 3 a schematic output power matrix and

Fig. 4 a schematic counting matrix.

In the drawings the same or similar elements and/or parts are labelled with the same reference numbers, so that no repeated explanation is necessary in each case.

Figure 1 shows a schematic flow diagram of a method according to the invention in one exemplary embodiment. The method for efficiency monitoring of a wind farm is started at 10. At 11, it is queried whether the reference time period has begun. If this question is answered with no, control returns to the connection between boxes 10 and 11. If this question is answered with "yes", in 12 the wind speed Vw, the wind direction Rw and the output power or output power value which is associated with this wind speed and wind direction, namely P (Vw, Rw) , is measured. At 13, the output power values are then stored in a first output power matrix, namely in a matrix element which corresponds to the measured wind speed and the measured wind direction, and in addition, the number 1 is added to the value in a first counting matrix in the matrix element which corresponds to this measured wind speed and the measured wind direction. Before the measurement the matrix elements are set to 0.

At 14 it is then queried whether the reference time period has ended. For a reference time period of one year, this is not yet the case. Therefore control returns to the point before box 12, where the wind speed, wind direction and the output power associated with this wind speed and this wind direction are measured. These values are then also stored in the first output power matrix and the first counting matrix. If the wind speed and wind direction have not changed, or not changed so much that these values must be saved in a different matrix element, the value 1 is added to this matrix element in the first counting matrix, so that a 2 is now there and in the first output power matrix the measured power is added. It can also be provided to store only averaged values in the first output power matrix, so that the two measurements that were made in the meantime are then added together and then divided by two. This averaging can be performed using any averaging methods, such as an arithmetic, geometric or another type of averaging.

Accordingly, in the course of the method a plurality of measured values can also occur per matrix element, the average value of which is then formed with regard to the output power values to be entered in the first output power matrix.

The measurement time, i.e. the time within which the wind speed, wind direction and the output power value relating to this wind speed and wind direction are measured, can be, for example, ten minutes. Here also, an average value of the corresponding measurements can be formed. To illustrate this, Fig. 3 shows a corresponding output power matrix and a corresponding counting matrix is shown in Fig. 4. This point will be discussed in further detail under the discussion of Figs. 3 and 4, however. If the reference time duration has ended, i.e. the query at 14 is answered with "yes", it is then queried whether the monitoring time period at 15 has begun. If the answer is no, control reverts to blocks 14 and 15, and if the answer is yes, processing continues with block 16, where again as in block 12 the wind speed Vw, the wind direction Rw and the output power value P (Vw, Rw) are measured and accordingly associated with this wind speed and this wind direction. The output power value is then stored in a second output power matrix and in a second counting matrix the value 1 is added to the matrix element which corresponds to the measured wind speed and the measured wind direction. At 18, it is then queried whether the monitoring time period has ended. If this is not the case, control reverts to block 16. If this is the case, the efficiency monitoring or an evaluation of the efficiency of the wind farm takes place in block 19. The final process step is shown schematically in Fig. 2 in a process flow chart.

In Fig. 2 the evaluation of the efficiency or the determination of the efficiency or the detection of an efficiency change of the wind farm begins at 20. In the next block at 21, the total reference output power RPMT is calculated via the above, for example, via the above-mentioned formula 1. At 22, the total monitoring output power is calculated according to the above Formula 2 as CPMT. At 23, the output power ratio is calculated according to the above formula 3 as PR. At 24 a query takes place, namely whether the output power ratio is less than or equal to a specifiable number. If this is not the case, the method must be terminated at 25 so that no further measures are to be taken and if this is answered in the affirmative, i.e. if the efficiency of the wind farm has decreased too greatly, at 26 appropriate measures must be evaluated to increase the efficiency again or to provide compensation.

In this case it can be provided that only those power values are taken into account in the calculation of the total reference output power and the total monitoring output power which have a sufficiently high counting value in the first counting matrix and the second counting matrix, in order to eliminate stochastic uncertainties. Missing matrix elements can be interpolated if necessary, or, if the uncertainties are too great, this range will be omitted from the evaluation.

Fig. 3 shows a schematic view of an output power matrix, which is divided into 25 wind strength sectors (wind speed). The wind velocities given there range from 0 m/s up to 24 m/s. In addition, the output power matrix is classified into wind direction ranges, here called sectors, namely into the wind direction ranges from 105° to 135° as a first range, 135° to 165° as a second range etc., until the range from 255° to 285°. The output powers specified in the matrix elements are average output powers in kW.

In the counting matrix shown schematically in Fig. 4, the same matrix elements or associations to the wind strength sectors and wind direction ranges are made. The numbers shown there indicate the frequency of a corresponding wind strength for a corresponding wind direction, or corresponding wind direction range .

All of the above-mentioned features, including those to be obtained from the drawings alone as well as individual features which are disclosed in combination with other features, are considered essential to the invention whether taken alone or in combination. Embodiments according to the invention can be fulfilled either by individual features or a combination of a plurality of features. Features that are marked with "in particular", are to be understood as optional features .

List of reference numerals 10 Start 11 Start reference time period? 12 Measure Vw, Rw, P (Vw, Rw) 13 Store in first output power matrix and in first counting matrix 14 Reference time period ended? 15 Start monitoring time period? 16 Measure Vw, Rw, P (Vw, Rw) 17 Store in second output power matrix and in second counting matrix 18 Monitoring time period ended? 19 Efficiency evaluation 20 Start efficiency evaluation

21 Calculate RPMT

22 Calculate CPMT

23 Calculate PR 24 PR < specifiable number? 25 End 26 Evaluate measures j Yes n No

Claims

1. Fremgangsmåde til effektivitetsovervågning af en vindenergianlægspark med følgende fremgangsmådetrin: måling (12) af i det mindste en vindhastighed og en vindretning over en forudfastsættelig tidsperiode, - måling (12) af en effekt af vindenergianlægsparken i den forudfastsættelige tidsperiode, - tilordning af effekten til den målte vindhastighed og den målte vindretning, lagring (13) af effekten som effektværdi i en første effektmatrix, i hvilke matrixelementer er tilordnet til forskellige parametre, omfattende i det mindste vindhastigheder og vindretninger, såfremt effekten udgør en repræsentativ værdi for vindenergianlægsparken, kendetegnet ved, at fremgangsmåden har følgende yderligere fremgangsmådetrin: - indretning af en første tællematrix med matrixelementer, der er tilordnet til de forskellige parametre, omfattende i det mindste vindhastigheder og vindretninger, addering af en tælleværdi til den første tællematrix' matrixelement, som svarer til den målte vindhastighed og den målte vindretning, hvor de ovenstående fremgangsmådetrin udføres for et multiplum af forudfastsættelige tidsperioder i en referencevarighed, - hvor der i en senere overvågningsvarighed udføres følgende fremgangsmådetrin for et multiplum af forudfastsættelige tidsperioder: måling (16) af i det mindste vindhastigheden og vindretningen over den forudfastsættelige tidsperiode, - måling (16) af en effekt af vindenergianlægsparken i den forudfastsættelige tidsperiode, - indretning af en anden tællematrix med matrixelementer, der tilordnes til de forskellige parametre, omfattende i det mindste vindhastigheder og vindretninger, addering af en tælleværdi til den anden tællematrix' matrixelement, som svarer til den målte vindhastighed og den målte vindretning, - hvor der med henblik på effektivitetsovervågningen sker en sammenligning af de med den anden tællematrix vægtede effektværdier i den første effektmatrix med summen af de under overvågningsvarigheden målte effekter.A method for efficiency monitoring of a wind energy park with the following method steps: measuring (12) at least one wind speed and wind direction over a predetermined time period, - measuring (12) an effect of the wind energy park during the predetermined time period, - assigning the power to the measured wind speed and the measured wind direction, storing (13) the power as a power value in a first power matrix, in which matrix elements are assigned to different parameters, including at least wind speeds and wind directions, if the power represents a representative value for the wind farm, characterized by, the method has the following additional process steps: - arranging a first counting matrix with matrix elements assigned to the various parameters, including at least wind speeds and wind directions, adding a count value to the first counting matrix's element corresponding to the n measured wind speed and measured wind direction, where the above process steps are performed for a plurality of predetermined time periods in a reference duration, - in a subsequent monitoring duration, the following process steps are performed for a multiple of predetermined time periods: measurement (16) of at least the wind speed and the wind direction over the predetermined time period, - measurement (16) of an effect of the wind energy park in the predetermined time period, - arrangement of another counting matrix with matrix elements assigned to the various parameters, including at least wind speeds and wind directions, adding a count value to the second element matrix matrix element corresponding to the measured wind speed and the measured wind direction - for the purpose of efficiency monitoring a comparison of the power values weighted with the second count matrix in the first power matrix with the sum of the below the waking duration measured effects.

2. Fremgangsmåde ifølge krav 1, kendetegnet ved, at fremgangsmådetrinene for et repræsentativt multiplum af forudfastsættelige tidsperioder gennemføres i referencevarigheden og overvågningsvarigheden.Method according to claim 1, characterized in that the process steps for a representative multiple of predetermined time periods are carried out in the reference duration and the monitoring duration.

3. Fremgangsmåde ifølge krav 1 eller 2, kendetegnet ved, at der under overvågningsvarigheden dannes en anden effektmatrix ved, at de i overvågningsvarigheden målte effekter tilordnes til de i den respektive tidsperiode målte vindhastigheder og vindretninger, og at effekterne lagres som effektværdier i den anden effektmatrix, idet der til effektværdierne i den anden effektmatrix er tilordnet forskellige parametre, omfattende i det mindste vindhastigheder og vindretninger.Method according to claim 1 or 2, characterized in that during the monitoring duration a different power matrix is formed by assigning the effects measured during the monitoring duration to the wind speeds and wind directions measured in the respective time period, and that the effects are stored as power values in the second power matrix. , having different parameters assigned to the power values in the second power matrix, including at least wind speeds and wind directions.

4. Fremgangsmåde ifølge et af kravene 1 til 3, kendetegnet ved, at der dannes en middelværdi af vindhastigheden og en middelværdi af vindretningen.Method according to one of claims 1 to 3, characterized in that a mean value of the wind speed and a mean value of the wind direction is formed.

5. Fremgangsmåde ifølge krav 3 eller 4, kendetegnet ved, at en middelværdi af effektværdierne lagres i de respektive matrixelementer i den første effektmatrix og den anden effektmatrix.Method according to claim 3 or 4, characterized in that a mean value of the power values is stored in the respective matrix elements in the first power matrix and the second power matrix.

6. Fremgangsmåde ifølge et af kravene 1 til 5, kendetegnet ved, at der til effektivitetsovervågningen kun tages hensyn til effektværdier, som i den tilordnede tællematrix har i det mindste n tælleværdier, hvor n er et naturligt tal, der er større end 1 og kan forudfastsættes.Method according to one of Claims 1 to 5, characterized in that the efficiency monitoring only takes into account power values which in the assigned count matrix have at least n count values, where n is a natural number greater than 1 and can be fixed in advance.

7. Fremgangsmåde ifølge et af kravene 3 til 6, kendetegnet ved, at der ved sammenligningen af de på den anden tællematrix normerede effektværdier i den første effektmatrix med summen af de under overvågningsvarigheden målte effekter dannes en sum af effektværdierne i den første effektmatrix multipliceret med den respektive tælleværdi i den anden tællematrix, hvor de to dannede summer derefter sammenlignes.Method according to one of claims 3 to 6, characterized in that in comparing the power values normed on the second count matrix with the sum of the effects measured during the monitoring duration, a sum of the power values in the first power matrix is multiplied by the respective count value in the second count matrix, where the two sums formed are then compared.

8. Fremgangsmåde ifølge et af kravene 3 til 7, kendetegnet ved, at summen af de under overvågningsvarigheden målte effekter fremkommer ved, at der dannes en sum af effektværdierne i den anden effektmatrix multipliceret med den respektive tælleværdi i den anden tællematrix.Method according to one of claims 3 to 7, characterized in that the sum of the effects measured during the monitoring duration is obtained by generating a sum of the power values in the second power matrix multiplied by the respective count value in the second count matrix.

9. Fremgangsmåde ifølge et af kravene 1 til 8, kendetegnet ved, at en effektivitetsændring først konstateres fra overskridelsen af en tolerancetærskel at regne.Method according to one of claims 1 to 8, characterized in that an efficiency change is first ascertained from the exceedance of a tolerance threshold to be calculated.

10. Fremgangsmåde ifølge et af kravene 1 til 9, kendetegnet ved, at der som yderligere parameter måles og tages hensyn til en vindgradient, en vindvariation, en vindturbulens, et atmosfærisk tryk, en luftfugtighed og/eller en temperatur af omgivelserne.Process according to one of Claims 1 to 9, characterized in that, as a further parameter, a wind gradient, a wind variation, a wind turbulence, an atmospheric pressure, an air humidity and / or a temperature of the ambient are measured and taken into account as an additional parameter.

11. Fremgangsmåde ifølge et af kravene 1 til 10, kendetegnet ved, at der som yderligere parameter tages hensyn til et spændingsniveau i det elektriske net og/eller en tilførsel af reaktiv effekt.Method according to one of claims 1 to 10, characterized in that as a further parameter a voltage level is taken into account in the electrical grid and / or a supply of reactive power.

12. Fremgangsmåde ifølge krav 10 eller 11, kendetegnet ved, at til en yderligere parameter tilvejebringer den første effektmatrix, den anden effektmatrix, den første tællematrix und den anden tællematrix hver især en yderligere dimension.Method according to claim 10 or 11, characterized in that for a further parameter, the first power matrix, the second power matrix, the first count matrix and the second count matrix each provide an additional dimension.

13. Fremgangsmåde ifølge et af kravene 1 til 11, kendetegnet ved, at referencevarigheden og overvågningsvarigheden hver især er et fuldstændigt meteorologisk år eller et heltals multiplum af et meteorologisk år.Method according to one of claims 1 to 11, characterized in that the reference duration and the monitoring duration are each a complete meteorological year or an integer multiple of a meteorological year.

14. Fremgangsmåde ifølge et af kravene 1 til 13, kendetegnet ved, at måleværdierne for vindhastigheden og vindretningen er inddelt i forudfastsættelige områder.Method according to one of claims 1 to 13, characterized in that the measured values for the wind speed and the direction of wind are divided into predetermined areas.

15. Fremgangsmåde ifølge et af kravene 1 til 14, kendetegnet ved, at værdierne af de yderligere parametre er inddelt i forudfastsættelige områder.Method according to one of claims 1 to 14, characterized in that the values of the additional parameters are divided into predetermined ranges.